Genetic variation of the RASGRF1 regulatory region affects human hippocampus-dependent memory

Abstract

The guanine nucleotide exchange factor RASGRF1 is an important regulator of intracellular signaling and neural plasticity in the brain. RASGRF1-deficient mice exhibit a complex phenotype with learning deficits and ocular abnormalities. Also in humans, a genome-wide association study has identified the single nucleotide polymorphism (SNP) rs8027411 in the putative transcription regulatory region of RASGRF1 as a risk variant of myopia. Here we aimed to assess whether, in line with the RASGRF1 knockout mouse phenotype, rs8027411 might also be associated with human memory function. We performed computer-based neuropsychological learning experiments in two independent cohorts of young, healthy participants. Tests included the Verbal Learning and Memory Test (VLMT) and the logical memory section of the Wechsler Memory Scale (WMS). Two sub-cohorts additionally participated in functional magnetic resonance imaging (fMRI) studies of hippocampus function. 119 participants performed a novelty encoding task that had previously been shown to engage the hippocampus, and 63 subjects participated in a reward-related memory encoding study. RASGRF1 rs8027411 genotype was indeed associated with memory performance in an allele dosage-dependent manner, with carriers of the T allele (i.e., the myopia risk allele) showing better memory performance in the early encoding phase of the VLMT and in the recall phase of the WMS logical memory section. In fMRI, T allele carriers exhibited increased hippocampal activation during presentation of novel images and during encoding of pictures associated with monetary reward. Taken together, our results provide evidence for a role of the RASGRF1 gene locus in hippocampus-dependent memory and, along with the previous association with myopia, point toward pleitropic effects of RASGRF1 genetic variations on complex neural function in humans.

Keywords: RASGRF1; episodic memory; fMRI; genetic variation; hippocampus.

Figure 1

Memory effect of RASGRF1 in Verbal Learning and Memory (VLMT). Higher % values indicate better memory performance. (A) In the first cohort we observed a significant positive correlation between T allele count and recall performance in the second and third learning trials. (B) In the second cohort we found a positive correlation between the T allele count and recall performance in the second learning trial. ^*Significant at p < 0.05; ^**Significant at p < 0.01.

Figure 2

Memory effect of RASGRF1 in a logical memory task WMS). Higher % values indicate better memory performance. (A) In the first cohort correlation analyses revealed a significant correlation between the T allele count and performance in the 24-h delayed recall test and a trend-wise positive correlation between T allele count and immediate recall performance. (B) In the second cohort, T allele count was positively correlated with performance in the immediate recall test and in delayed recall after 30 min. ^*Significant at p < 0.05.

Figure 3

Hippocampal novelty processing and its modulation by RASGRF1 genotype (main effect of RASGRF1). TT carriers exhibited increased hippocampal activation, when compared to G allele carriers. This effect was significant at p < 0.017, FWE-corrected for ROI volume. Activations are superimposed on the MNI template brain provided by MRIcron. Coordinates are in MNI space. Bar plots depict contrasts of parameter estimates at the peak coordinate separated by genotype. Error bars depict standard errors of the mean.

Figure 4

The hippocampal recognition-encoding response for monetary reward-predicting items and its modulation by RASGRF1 genotype [constrast TT vs. G (GT+GG) masked (incl.) with the main effect of RASGRF1 at p < 0.05]. TT carriers showed highest hippocampal activation, when compared to G allele carries (p = 0.009, FWE-corrected for ROI volume). Activations are superimposed on the MNI template brain provided by MRIcron. Coordinates are in MNI space. Bar plots depict contrasts of parameter estimates at the peak coordinate separated by genotype. Error bars depict standard errors of the mean.